HIV-1 transmission through oral and gastrointestinal tissues is a well-known route of infection in breast-feeding infants. The tonsils is considered a potential portal of viral entry. Little is known of the cell types infected in tonsils following breast milk ingestion, viral determinants of transmission in the nursing infant, and innate inhibitors that protect oral tissues against infection. Efforts to understand the mechanisms of breastfeeding transmission have been limited by the lack of a tissue-based model. This exploratory study will develop and characterize a human tonsil histoculture model of oral HIV-1 acquisition through breast milk. AIM 1: Human tonsil histocultures will be established and infected with clinical HIV-1 isolates obtained from breast milk and perinatally infected infants, including babies infected via breastfeeding. Virus production in histocultures will be assessed by p24 ELISA, and infected cell types will be identified by flow cytometry and immunohistochemistry. Results will be compared with histoculture infections of prototypic R5, X4 and R5X4 viruses. AIM 2: To identify the genetic and phenotypic properties of HIV-1 variants productively infecting oral tissues, recombinant infectious HIV-1 vectors containing the V1 -V3 gp120 regions of breast milk-derived env sequences will be generated, evaluated for syncytium-inducing phenotype (MT-2 assay) and coreceptor usage pattern (beta-gal infectivity assay), and tested in the model. The diversity of recovered viruses will be assessed in V1 -V3 heteroduplex tracking assays and genotyping. Results will be compared to vectors unable to establish productive infection and to the isolates described above. These experiments will test the hypothesis that an R5 non-syncytium-inducing major variant is preferentially transmitted from milk to oral tissue. AIM 3: To test the hypothesis that a subset of innate inhibitors provides the greatest protection of oral tissue several anti-HIV-1 salivary proteins previously identified in cell models will be tested, singly and in combination, in the tonsil model using the HIV-1 env vectors and isolates. Breast milk and saliva from infected and uninfected women will also be tested to determine the impact of maternal antibodies on oral HIV-1 acquisition. The proposed studies will enhance our understanding of postnatal transmission and oral infection by defining viral determinants for future vaccine development and identifying innate inhibitors that might serve as novel agents for disrupting breastfeeding transmission.